Unravelling the Complexity: Understanding the Challenges of Reinforcement Learning
Abstract: After an extensive research and exploration of supervised, unsupervised and semi-supervised machine learning algorithms, researchers across the numerous application domains of machine learning are now looking to implement reinforcement learning techniques as they promise a realization of more human-like intelligence in machines. This paper presents a comprehensive body of knowledge about the complexities and challenges that researchers might face while developing reinforcement learning models as solutions for real-life problems. Also, some recommendations have been made in order to assist effective implementation of reinforcement learning algorithms.
Keywords: computational complexity, environment specification, exploration-exploitation, reinforcement learning, safeRL, sample efficiency
How to Cite?: Brahmaleen Kaur Sidhu, "Unravelling the Complexity: Understanding the Challenges of Reinforcement Learning", Volume 13 Issue 3, March 2024, International Journal of Science and Research (IJSR), Pages: 233-239, https://www.ijsr.net/getabstract.php?paperid=SR24304190409, DOI: https://dx.doi.org/10.21275/SR24304190409